Re: [PATCH 0/2] platform/x86: lenovo-wmi-capdata: Add debugfs file for dumping capdata

[Derek J. Clark]

On February 11, 2026 3:49:43 AM PST, Rong Zhang <i@xxxxxxxx> wrote:
>Hi Derek and Kurt,
>
>On Tue, 2026-02-10 at 15:49 -0800, Derek J. Clark wrote:
>> On February 10, 2026 3:41:31 PM PST, Kurt Borja <kuurtb@xxxxxxxxx> wrote:
>> > On Tue Feb 10, 2026 at 2:19 PM -05, Rong Zhang wrote:
>> > > The Lenovo GameZone/Other interfaces have some delicate divergences
>> > > among different devices. When making a bug report or adding support for
>> > > new devices/interfaces, capdata is the most important information to
>> > > cross-check with.
>> > > 
>> > > Add a debugfs file (lenovo_wmi/<device_name>/capdata), so that users can
>> > > dump capdata and include it in their reports.
>> > > 
>> > > The output is like:
>> > > 
>> > >   LENOVO_CAPABILITY_DATA_00[0]:
>> > >     id:             0x00010000 [dev:  0, feat:  1, mode:  0, type:  0]
>> > >     supported:      0x00000007 [ RWV]
>> > >     default_value:  0
>> > > 
>> > >   LENOVO_CAPABILITY_DATA_01[0]:
>> > >     id:             0x00000000 [dev:  0, feat:  0, mode:  0, type:  0]
>> > >     supported:      0x00000000 [    ]
>> > >     default_value:  0
>> > >     step:           0
>> > >     min_value:      0
>> > >     max_value:      0
>> > > 
>> > >   LENOVO_FAN_TEST_DATA[0]:
>> > >     id:             1
>> > >     min_rpm:        2200
>> > >     max_rpm:        5000
>> > > 
>> > > A helper function for creating per-devcie debugfs directories is also
>> > > introduced into lenovo-wmi-helpers in order that we can maintain a tidy
>> > > directory structure in debugfs.
>> > > 
>> > > The series is based on platform-drivers-x86/review-ilpo-next since it
>> > > depends on a commit there. I am OK to wait until the next cycle.
>> > > 
>> > > Rong Zhang (2):
>> > >   platform/x86: lenovo-wmi-helpers: Add helper for creating per-device
>> > >     debugfs dir
>> > >   platform/x86: lenovo-wmi-capdata: Add debugfs file for dumping capdata
>> > > 
>> > >  drivers/platform/x86/lenovo/Kconfig       |   1 +
>> > >  drivers/platform/x86/lenovo/wmi-capdata.c | 121 ++++++++++++++++++++++
>> > >  drivers/platform/x86/lenovo/wmi-capdata.h |   4 +-
>> > >  drivers/platform/x86/lenovo/wmi-helpers.c |  34 ++++++
>> > >  drivers/platform/x86/lenovo/wmi-helpers.h |   2 +
>> > >  5 files changed, 159 insertions(+), 3 deletions(-)
>> > > 
>> > > 
>> > > base-commit: 5a5203a45b063a594e89a2aeaf9e4923893a5b4c
>> > 
>> > Hi Rong,
>> > 
>> > This is indeed very useful debug information.
>> > 
>> > Once you and Derek agree on the approach for this series, I'll add my
>> > Tested-by tag.
>> > 
>> > One more thing though. Can you please explain the "supported" byte? What
>> > is `+` and `V` supposed to mean?
>
>"V" means valid. It's consistent with the corresponding macro name.
>
>> Kurt,
>> 
>> It's enumerated in the patch:
>> 
>> +       seq_printf(s, "  supported:      0x%08x [%c%c%c%c]\n", cd00->supported,
>> +                  extra ? '+' : ' ',
>> +                  get   ? 'R' : ' ',
>> +                  set   ? 'W' : ' ',
>> +                  valid ? 'V' : ' ');
>> 
>> 
>> From the documentation:
>> uint32 Capability //7:by project 
>> bit 2:  0: not support SetFeatureValue(),  1: support SetFeatureValue() 
>> bit 1:  0: not support GetFeatureValue(),  1: support GetFeatureValue()
>> bit 0:  0: not support feature,  1: support feature 
>> 
>> I don't see any examples of the "extra" bit in the docs.
>
>It's not a single bit. Instead, it's printed out when any bit other
>than RWV is set. I added this when I recalled the definition of
>capability 0x04050000, which has an extra bit defined:
>
>   Fan Test For Diagnostic Software
>   uint32 IDs //0x04050000
>   uint32 Capability //9:by project
>   bit 3: 0: not support LENOVO_FAN_TEST_DATA, 1 support LENOVO_FAN_TEST_DATA
>   bit 2: 0: not support SetFeatureValue(), 1: support SetFeatureValue()
>   bit 1: 0: not support GetFeatureValue(), 1: support GetFeatureValue()
>   bit 0: 0: not support fan test for diagnostic software, 1: support an test for diagnostic software
>
>(from https://lore.kernel.org/r/CAFqHKTkOZUfDb8cGbGnVPCS9wNbOBsiyOk_MkZR-2_Za6ZPMng@xxxxxxxxxxxxxx/ )
>
>I assume extra bits are not standardized, and some (current or future)
>capabilities may have multiple extra bits defined. Hence, I simply
>named it "+". After all, it's just a reminder for us to check the
>documentation (I don't have any, though).

Okay. Makes sense. I have no issues with reporting that then. 

Thanks,
Derek

>Thanks,
>Rong
>
>> Thanks,
>> Derek